Polymer polyols, as that term is used herein, refers to polyvinyl polymer dispersions prepared by the in situ polymerization of one or more vinyl monomers in a polyoxyalkylene "base" polyol. Polymer-modified polyols, as that term is used herein, refers to polyoxyalkylene polyether polyols having a dispersed phase of a urea or urethane/urea polymer prepared by the in situ polymerization of a diisocyanate or polyisocyanate with an isocyanate-reactive monomer, preferably an amino-functional monomer such as an alkanolamine, diamine, or the like. The majority of such polymer polyols and polymer-modified polyols are used in the polyurethane field for diverse applications, including cell openers and hardness enhancers for polyurethane foam, and as reinforcing additives for a variety of microcellular and non-cellular polyurethanes.
The manufacture of polymer polyols is by now well known, and may involve batch, semi-batch, and fully continuous processes. In all of these processes, one or more vinyl monomers such as acrylonitrile and styrene are polymerized in situ in one or more base polyols, with or without the presence of an added stabilizer. The amount of monomer(s) fed to the reactor is selected to achieve the desired vinyl polymer solids content in the final polymer polyol product. The solids level may range from as little as 5 weight percent to upwards of 60 weight percent, however, it is most economical to produce polymer polyols at relatively high solids loadings even when a low solids product is desired. If a lower solids content polymer polyol is desired, the solids content may be lowered by dilution of the higher solids polyol with further amounts of the same base polyol or other non-polymer polyol, or by blending with a polymer polyol of lesser solids content. The base polyol functionality is dictated by the particular polyurethane end-use desired, and may typically involve nominal functionalities of two to eight. The details of polymer polyol manufacture will be presented hereafter.
The manufacture of polymer-modified polyols is also by now well known. The two most common polymer-modified polyols are the so-called PIPA (Polylsocyanate PolyAddition) polyols and the PHD (PolyHarnstoff Dispersion) polyols. Both these polymer-modified polyols and others are prepared by the addition polymerization of an isocyanate, for example a di- or polyisocyanate, with an isocyanate-reactive monomer, preferably an amino-functional compound: an alkanolamine in the case of PIPA polyols, and a di- or polyamine in the case of PHD polyols. Mixtures of these isocyanate reactive monomers as well as reactive diols may also be used. The reactive monomers are polymerized in situ in a polyoxyalkylene polyether polyol which forms the continuous phase of the polymer-modified polyol. In many cases, a portion of the polyol continuous phase becomes associated with the polymer phase by reaction with isocyanate groups. More detailed description of polymer-modified polyols is presented hereinafter.
In both polymer polyols and polymer-modified polyols, the monomers are generally initially soluble in the polyol continuous phase, as are in general the initial low molecular weight oligomers. However, as the molecular weight of the polymer phase grows, the polymer becomes insoluble, forming small particles which rapidly coalesce and/or agglomerate to larger particles in the submicron to several micron range. Hereinafter, the term "polymer polyol" will refer to dispersions of vinyl polymers, "polymer-modified polyol" to polyurea, polyurethaneurea, or other isocyanate-derived polymer dispersions, and the term "polyol polymer dispersions" will refer to both of these collectively.
The base polyols used in preparing polyol polymer dispersions generally contain a high proportion of polyoxypropylene moieties. Polyoxypropylene polyether polyols are conventionally prepared by the base-catalyzed oxyalkylation of a suitably functional initiator molecule with propylene oxide or a mixture of propylene oxide and ethylene oxide. During base-catalyzed oxypropylation, a competing rearrangement of propylene oxide into allyl alcohol continually introduces this unsaturated monol into the polymerization reactor. The allyl alcohol acts as an additional initiator, and being monofunctional, lowers the actual functionality of the polyol. The continued creation of low molecular weight monofunctional species also broadens the molecular weight distribution. As a result of these effects, the practical upper limit of polyoxypropylene polyether polyols equivalent weight is c.a. 2000 Da (Daltons).
For example, a 4000 Da molecular weight base-catalyzed polyoxypropylene diol may contain 0.07 to 0.12 meq. unsaturation per gram polyol, amounting to from 25-40 mol percent of monol. As a result, the polyol nominal functionality of two is reduced to actual functionalities of c.a. 1.6 to 1.7 or less. Unsaturation is generally measured in accordance with ASTM test D-2849-69 "Testing Urethane Foam Polyol Raw Materials."
Lowering the oxypropylation temperature and decreasing the amount of basic catalyst allows for some reduction of unsaturation, but at the expense of greatly extended reaction time which is not commercially acceptable. Moreover, the reduction in unsaturation is but slight. Use of alternative catalyst systems, for example cesium hydroxide rather than the more commonly used sodium or potassium hydroxides; strontium or barium hydroxides; dialkyl zinc; metal naphthenates; and combinations of metal naphthanates and tertiary amines have all been proposed. However, the unsaturation is generally reduced only to about 0.03 to 0.04 meq/g by these methods, still representing 10-15 mol percent monol. In all these cases, the catalyst residues must be removed prior to the in situ polymerization of vinyl or other monomers to produce polyol polymer dispersions. Basic catalysts are generally removed by adsorption with magnesium silicate followed by filtration, by neutralization followed by filtration, or through the use of ion-exchange techniques.
In the 1960's, double metal cyanide catalysts such as complexes of zinc hexacyanocobaltate were found to be useful in a variety of polymerization reactions, as evidenced by U.S. Pat. Nos. 3,427,256, 3,427,334, 3,427,335, 3,829,505, 3,941,849, and 4,242,490. In polymerization of propylene oxide, such catalysts were found to produce polyols with unsaturation in the range of 0.02 meq./g. However, even though relatively active catalysts, their cost relative to activity was quite high. In addition, catalyst removal was problematic. Refinements in double metal cyanide complex catalysts have led to catalysts with somewhat higher activity, as evidenced by U.S. Pat. Nos. 4,472,560, 4,477,589, 4,985,491, 5,100,997, and 5,158,922. These catalysts, generally glyme complexes of zinc hexacyanocobaltate, were effective in preparing polyoxypropylene polyols with unsaturation levels of c.a. 0.015 to 0.018 meq/g. Despite being more active than the prior catalysts, the cost of these improved catalysts, in addition to the difficulties associated with catalyst removal, again prevented any large scale commercialization.
Recently, however, exceptionally active double metal cyanide complex catalysts have been developed at the ARCO Chemical Co., as evidenced by copending U.S. application Ser. No. 08/156,534 now U.S. Pat. No. 5,470,813 and application Ser. No. 08/302,296 now U.S. Pat. No. 5,482,908, herein incorporated by reference. In addition to their much higher activity as compared to previous double metal cyanide complex catalysts, these catalysts have further been shown suitable for producing polyoxypropylene polyols with measured unsaturation in the range of 0.003 to 0.007 meq/g. Not only is the measured unsaturation exceptionally low, but moreover, despite the fact that unsaturation is generally accepted as a measure of monol content, lower molecular weight species are not detected by gel permeation chromatography. The polyoxypropylene polyols are truly monodisperse, having a very narrow molecular weight distribution. Despite being much more active catalysts than prior catalysts and being more susceptible to simple filtration for catalyst removal, the necessity to finely filter or otherwise remove catalyst residues prior to use as base polyols for polyol polymer dispersion production undesirably increases processing time.
In Japanese published application H2-294319 (1990), double metal cyanide complex catalysts were used to prepare polyoxypropylene polyols following which the double metal cyanide catalyst residues were denatured by adding alkali metal hydroxide which then served as the oxyalkylation catalyst for capping the polyoxypropylene polyols with oxyethylene moieties. Following removal of the catalyst residues, high primary hydroxyl content polymer polyols were prepared in a conventional manner. Similar polymer polyols prepared by in situ polymerization in oxyethylene capped polyoxypropylene polyols are disclosed in U.S. Pat. Nos. 5,093,380 and 5,300,535.
In Japanese published application 5-39428 (1993), unspecific zinc hexacyanocobaltate catalysts were used to prepare polyoxypropylene polyols which were then used as base polyols for polymer polyol manufacture, with or without further addition of double metal cyanide catalyst as a vinyl polymerization catalyst. However, the presence of large amounts of double metal cyanide catalyst residues in the polymer polyol product, even if they did not affect subsequent in situ vinyl polymerization, is undesirable. In the food processing industry and medical prostheses industries, for example, heavy metal ion content must be minimal.
J. L. Schuchardt and S. D. Harper, "Preparation Of High Molecular Weight Polyols Using Double Metal Cyanide Catalysts, " 32ND ANNUAL POLYURETHANE TECNIAL MARKETING CONFERENCE, Oct. 1-4, 1989, discloses that double metal cyanide complex catalyst residues can increase the viscosity of isocyanate-terminated prepolymers prepared from polyols containing such residues, this viscosity increase believed due to allophanate formation. Herrold et al. in U.S. Pat. No. 4,355,188 and the many other patents directed to removal of catalyst residues, e.g., U.S. Pat. Nos. 3,427,256, 5,248,833, 4,721,818, 5,010,047, and 4,987,271 attest to the commercial significance of double metal cyanide catalyst removal.
It would be desirable to provide a method of preparing polyol polymer dispersions from double metal cyanide catalyzed pqlyoxypropylene polyether polyols without the necessity of removing or denaturing double metal cyanide complex catalyst residues, without such catalyst residues appearing in the continuous polyol phase of the polyol polymer dispersion. It would be further desirable to prepare polymer polyols which are white or off-white in color.